Extropic Develops Thermodynamic Computing Hardware for Energy-Efficient AI Processing
By
vyrotek
Fresh out the oven, still warm. Top of the tray.
Summary
Extropic is developing thermodynamic computing hardware designed to be significantly more energy efficient than GPUs for AI workloads. The company has created thermodynamic sampling units (TSUs) that are inherently probabilistic, making them suitable for probabilistic AI applications. They've developed the XTR-0 prototype platform that enables low-latency communication between Extropic chips and traditional processors, and have released THRML, an open-source Python library to support development on their hardware platform.
Key quotes
· 4 pulledExtropic is building thermodynamic computing hardware that is radically more energy efficient than GPUs.
Our thermodynamic sampling units (TSUs) are inherently probabilistic, the perfect fit for probabilistic AI workloads.
XTR-0 enables the development of ultra-efficient AI algorithms by providing low-latency communication between Extropic chips and a traditional processor.
INTRODUCING THRML: Our open-source Python library that enables development on thermodynamic hardware.
You might also wanna read
Morgan Stanley: AI capabilities, not price, will drive China's EV market competition
Morgan Stanley analysts predict that competition in China's EV market is shifting from price wars to AI capabilities, particularly autonomou
scmp.com·4h ago
AI boom drives energy sector transformation as electricity becomes strategic business asset
The AI boom is driving a massive scramble for electricity, transforming energy from a cheap commodity into a strategic business asset. Compa
axios.com·8h ago
Rivian argues AI-powered assistants will make Apple CarPlay obsolete
Rivian defends its decision to exclude Apple CarPlay and Android Auto from its vehicles, arguing that AI-powered in-car assistants will soon
digitaltrends.com·9h ago
How AI Company IPOs Could Spark a New Wave of Philanthropy
The article discusses how the wave of AI company IPOs is creating a new class of tech billionaires, and explores what these fortunes might m
nytimes.com·9h agoHow AI Company IPOs Could Spark a New Wave of Philanthropy
The article discusses how the wave of AI company IPOs is creating a new class of tech billionaires, and explores what these fortunes might m
nytimes.com·9h ago
Sergey Brin tells Google AI staff 60-hour workweek is the 'sweet spot' for productivity
Sergey Brin, Google cofounder, wrote an internal memo to employees working on the Gemini AI products recommending a 60-hour workweek as the
fortune.com·9h ago